Skip to content

Commit a48c85a

Browse files
brettinanlbrettinanl
committed
resolved merge conflicts with release_04 branch
2 parents d6c8450 + c196441 commit a48c85a

File tree

304 files changed

+1957968
-8363
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

304 files changed

+1957968
-8363
lines changed

‎.pre-commit-config.yaml

+7
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,7 @@
1+
# See https://pre-commit.com for more information
2+
# See https://pre-commit.com/hooks.html for more hooks
3+
repos:
4+
- repo: https://github.com/pycqa/flake8
5+
rev: '3.9.2' # pick a git hash / tag to point to
6+
hooks:
7+
- id: flake8

‎Pilot1/Attn/attn.py

+51-63
Original file line numberDiff line numberDiff line change
@@ -7,13 +7,7 @@
77
import pandas as pd
88
import numpy as np
99

10-
from sklearn.metrics import mean_squared_error
11-
from sklearn.metrics import r2_score
12-
from scipy.stats.stats import pearsonr
13-
1410
file_path = os.path.dirname(os.path.realpath(__file__))
15-
#lib_path = os.path.abspath(os.path.join(file_path, '..'))
16-
#sys.path.append(lib_path)
1711
lib_path2 = os.path.abspath(os.path.join(file_path, '..', '..', 'common'))
1812
sys.path.append(lib_path2)
1913

@@ -22,45 +16,41 @@
2216
logger = logging.getLogger(__name__)
2317
candle.set_parallelism_threads()
2418

25-
additional_definitions = [
26-
{'name':'latent_dim',
27-
'action':'store',
28-
'type': int,
29-
'help':'latent dimensions'},
30-
{'name':'residual',
31-
'type': candle.str2bool,
32-
'default': False,
33-
'help':'add skip connections to the layers'},
34-
{'name':'reduce_lr',
35-
'type': candle.str2bool,
36-
'default': False,
37-
'help':'reduce learning rate on plateau'},
38-
{'name':'warmup_lr',
39-
'type': candle.str2bool,
40-
'default': False,
41-
'help':'gradually increase learning rate on start'},
42-
{'name':'base_lr',
43-
'type': float,
44-
'help':'base learning rate'},
45-
{'name':'epsilon_std',
46-
'type': float,
47-
'help':'epsilon std for sampling latent noise'},
48-
{'name':'use_cp',
49-
'type': candle.str2bool,
50-
'default': False,
51-
'help':'checkpoint models with best val_loss'},
52-
#{'name':'shuffle',
53-
#'type': candle.str2bool,
54-
#'default': False,
55-
#'help':'shuffle data'},
56-
{'name':'use_tb',
57-
'type': candle.str2bool,
58-
'default': False,
59-
'help':'use tensorboard'},
60-
{'name':'tsne',
61-
'type': candle.str2bool,
62-
'default': False,
63-
'help':'generate tsne plot of the latent representation'}
19+
additional_definitions = [
20+
{'name': 'latent_dim',
21+
'action': 'store',
22+
'type': int,
23+
'help': 'latent dimensions'},
24+
{'name': 'residual',
25+
'type': candle.str2bool,
26+
'default': False,
27+
'help': 'add skip connections to the layers'},
28+
{'name': 'reduce_lr',
29+
'type': candle.str2bool,
30+
'default': False,
31+
'help': 'reduce learning rate on plateau'},
32+
{'name': 'warmup_lr',
33+
'type': candle.str2bool,
34+
'default': False,
35+
'help': 'gradually increase learning rate on start'},
36+
{'name': 'base_lr',
37+
'type': float,
38+
'help': 'base learning rate'},
39+
{'name': 'epsilon_std',
40+
'type': float,
41+
'help': 'epsilon std for sampling latent noise'},
42+
{'name': 'use_cp',
43+
'type': candle.str2bool,
44+
'default': False,
45+
'help': 'checkpoint models with best val_loss'},
46+
{'name': 'use_tb',
47+
'type': candle.str2bool,
48+
'default': False,
49+
'help': 'use tensorboard'},
50+
{'name': 'tsne',
51+
'type': candle.str2bool,
52+
'default': False,
53+
'help': 'generate tsne plot of the latent representation'}
6454
]
6555

6656
required = [
@@ -80,7 +70,8 @@
8070
'batch_normalization',
8171
'epsilon_std',
8272
'timeout'
83-
]
73+
]
74+
8475

8576
class BenchmarkAttn(candle.Benchmark):
8677

@@ -102,7 +93,7 @@ def extension_from_parameters(params, framework=''):
10293
ext = framework
10394
for i, n in enumerate(params['dense']):
10495
if n:
105-
ext += '.D{}={}'.format(i+1, n)
96+
ext += '.D{}={}'.format(i + 1, n)
10697
ext += '.A={}'.format(params['activation'][0])
10798
ext += '.B={}'.format(params['batch_size'])
10899
ext += '.E={}'.format(params['epochs'])
@@ -124,15 +115,17 @@ def extension_from_parameters(params, framework=''):
124115
ext += '.Res'
125116

126117
return ext
118+
119+
127120
def load_data(params, seed):
128121

129122
# start change #
130123
if params['train_data'].endswith('h5') or params['train_data'].endswith('hdf5'):
131-
print ('processing h5 in file {}'.format(params['train_data']))
124+
print('processing h5 in file {}'.format(params['train_data']))
132125

133126
url = params['data_url']
134127
file_train = params['train_data']
135-
train_file = candle.get_file(file_train, url+file_train, cache_subdir='Pilot1')
128+
train_file = candle.get_file(file_train, url + file_train, cache_subdir='Pilot1')
136129

137130
df_x_train_0 = pd.read_hdf(train_file, 'x_train_0').astype(np.float32)
138131
df_x_train_1 = pd.read_hdf(train_file, 'x_train_1').astype(np.float32)
@@ -152,7 +145,7 @@ def load_data(params, seed):
152145
Y_train = pd.read_hdf(train_file, 'y_train')
153146
Y_test = pd.read_hdf(train_file, 'y_test')
154147
Y_val = pd.read_hdf(train_file, 'y_val')
155-
148+
156149
# assumes AUC is in the third column at index 2
157150
# df_y = df['AUC'].astype('int')
158151
# df_x = df.iloc[:,3:].astype(np.float32)
@@ -161,18 +154,17 @@ def load_data(params, seed):
161154
# scaler = StandardScaler()
162155
# df_x = scaler.fit_transform(df_x)
163156
else:
164-
print ('expecting in file file suffix h5')
157+
print('expecting in file file suffix h5')
165158
sys.exit()
166-
167-
159+
168160
print('x_train shape:', X_train.shape)
169161
print('x_test shape:', X_test.shape)
170-
162+
171163
return X_train, Y_train, X_val, Y_val, X_test, Y_test
172164

173165
# start change #
174166
if train_file.endswith('h5') or train_file.endswith('hdf5'):
175-
print ('processing h5 in file {}'.format(train_file))
167+
print('processing h5 in file {}'.format(train_file))
176168

177169
df_x_train_0 = pd.read_hdf(train_file, 'x_train_0').astype(np.float32)
178170
df_x_train_1 = pd.read_hdf(train_file, 'x_train_1').astype(np.float32)
@@ -192,23 +184,19 @@ def load_data(params, seed):
192184
Y_train = pd.read_hdf(train_file, 'y_train')
193185
Y_test = pd.read_hdf(train_file, 'y_test')
194186
Y_val = pd.read_hdf(train_file, 'y_val')
195-
187+
196188
# assumes AUC is in the third column at index 2
197189
# df_y = df['AUC'].astype('int')
198190
# df_x = df.iloc[:,3:].astype(np.float32)
199191

200192
# assumes dataframe has already been scaled
201193
# scaler = StandardScaler()
202194
# df_x = scaler.fit_transform(df_x)
203-
204195
else:
205-
print ('expecting in file file suffix h5')
196+
print('expecting in file file suffix h5')
206197
sys.exit()
207-
208-
198+
209199
print('x_train shape:', X_train.shape)
210200
print('x_test shape:', X_test.shape)
211-
212-
return X_train, Y_train, X_val, Y_val, X_test, Y_test
213-
214201

202+
return X_train, Y_train, X_val, Y_val, X_test, Y_test

‎Pilot1/Attn/attn_abs_default_model.txt

+3-2
Original file line numberDiff line numberDiff line change
@@ -4,7 +4,7 @@ train_data='top_21_1fold_001.h5'
44
model_name='attn_abs'
55
dense=[1000, 1000, 1000, 500, 250, 125, 60, 30, 2]
66
batch_size=32
7-
epochs=2
7+
epochs=50
88
activation=['relu', 'relu', 'softmax', 'relu', 'relu', 'relu', 'relu', 'relu', 'softmax']
99
loss='categorical_crossentropy'
1010
optimizer='sgd'
@@ -21,7 +21,8 @@ output_dir='save_abs/EXP01/'
2121
experiment_id='01'
2222
run_id='1'
2323
save_path='save_abs/EXP01/'
24-
target_abs_acc=0.85
24+
min_abs_acc=0.85
25+
max_abs_frac=0.3
2526

2627
[Monitor_Params]
2728
timeout=3600

0 commit comments

Comments
 (0)